ABSTRACT
Microarray data provide lots of information regarding gene expression levels. Due to the large amount of such data, their analysis requires sufficient computational methods for identifying and analyzing gene regulation networks; however, researchers in this field are faced with numerous challenges such as consideration for too many genes and at the same time, the limited number of samples and their noisy nature of the data. In this paper, a hybrid method base on fuzzy cognitive map and compressed sensing is used to identify interactions between genes. For this purpose, in inference of the gene regulation network, the Ensemble Kalman filtered compressed sensing is used to learn the fuzzy cognitive map. Using the Ensemble Kalman filter and compressed sensing, the fuzzy cognitive map will be robust against noise. The proposed algorithm is evaluated using several metrics and compared with several well know methods such as LASSOFCM, KFRegular, CMI2NI. The experimental results show that the proposed method outperforms methods proposed in recent years in terms of SSmean, Data Error and accuracy.
ABSTRACT
OBJECTIVES: The purpose of our study was to assess the value of aVR ST-segment elevation (STE) during acute non ST-segment elevation myocardial infarction (NSTEMI) or unstable angina. BACKGROUND: STE in lead aVR has been associated with severe coronary lesions in patients with acute coronary syndromes. However, there are conflicting data regarding the prognostic significance of this finding. METHODS: We evaluated the initial electrocardiogram (ECG) in 129 patients admitted to our hospital with acute NSTEMI or unstable angina without STE in leads other than aVR or V1. STE in aVR lead was measured and echocardiography and coronary angiography were performed within 48-72 hours after hospitalization. RESULTS: Overall, 40.3% (52 patients) had more than 0.05 mv STE in lead aVR. These patients had an increased prevalence of ST ≥ 1 mm in lead V1, a more frequent and extensive ST-segment depression (STD) in other leads, a higher prevalence of anterior and lateral STD and a lower frequency of isolated negative T waves. It was also strongly associated with cardiac enzyme rising and a trend toward higher 3-month mortality. Furthermore, patients with STE in lead aVR were more likely to have three-vessel or multivessel disease, higher Gensini score of the coronary arteries, lower left ventricular ejection fraction (LVEF) and higher incidence of mitral regurgitation (MR). CONCLUSIONS: Our study showed that among ECG markers, the sole criterion STE in lead aVR was independently associated with atherosclerosis severity and decreased LVEF. Also, it was significantly associated with the presence of MR.